Presentation is loading. Please wait.

Presentation is loading. Please wait.

Using Predictive Modeling to Target Value-Based Prescription Management and Pharmacy Benefit Efficiency Pat Gleason, PharmD, BCPS, FCCP Director of Outcomes.

Similar presentations


Presentation on theme: "Using Predictive Modeling to Target Value-Based Prescription Management and Pharmacy Benefit Efficiency Pat Gleason, PharmD, BCPS, FCCP Director of Outcomes."— Presentation transcript:

1 Using Predictive Modeling to Target Value-Based Prescription Management and Pharmacy Benefit Efficiency Pat Gleason, PharmD, BCPS, FCCP Director of Outcomes Assessment and Adjunct Associate Professor University of Minnesota, College of Pharmacy

2 © PRIME THERAPEUTICS LLC 2 Prime Therapeutics ●Pharmacy Benefits Manager (PBM) ●Privately held organization owned by 10 BCBS plans (FL, IL, KS, MN, NE, NM, ND, OK, TX, and WY) ●14.6 million covered lives; 36 million weighted lives across 16 different BCBS plans (coast-to-coast)  900,000 Medicare Part D lives  9.6 million full service PBM lives  Claims processing, clinical and medication management, formulary, rebate administration, mail order, specialty pharmacy, national pharmacy network ●Integrate medical and pharmacy claims data for total health care management

3 © PRIME THERAPEUTICS LLC 3 Health Expenditures & Quality ●National Health Expenditures (NHE) are increasing 1  Health care cost trends are expected to continue around 7% through 2017 outpacing economic growth  Pharmacy spending has been increasing even more quickly than medical spending in the U.S., projected to be 11% of NHE by 2012 ●Quality of health care is decreasing 2  U.S. ranked 37 in the world in health care performance  Only 55% of Americans receive the care they need, as indicated by medical evidence  79,000 Americans die each year because they do not receive evidence-based care for chronic conditions 1) Keehan S, et al. Health Spending Projections Through 2017. Health Affairs 2008:27:w145-w155. 2) Boehm, Jennifer. “The Road Ahead: Value-Based Design.” Hewitt Associates. 1 May 2007.

4 © PRIME THERAPEUTICS LLC 4 Kaiser 2007 Employer Survey

5 © PRIME THERAPEUTICS LLC 5 Kaiser 2007 Employer Survey: Prescription Copays

6 © PRIME THERAPEUTICS LLC 6 Value-Based Insurance Design in Action ●Companies like Pitney Bowes and Marriott International have reduced or eliminated copayments for all users of drugs for chronic conditions, such as diabetes, asthma and hypertension 1,2 ●Consultants are promoting a valued-based approach  The Hewitt Value Report Card requires insurers and PBMs to assess utilization of key drug classes and ensure self-funded groups of the efficient use of medications in each drug class 1) Freudenheim, Milt. “To Save Later, Some Employers are Offering Free Drugs Now.” The New York Times. 21 February 2007. 2) Mahoney JJ. Reducing Patient Drug Acquisition Costs Can Lower Diabetes Health Claims. Am J Manag Care 2005;11:S170-S176

7 © PRIME THERAPEUTICS LLC 7 Value-Based Insurance Design ●Current implementation of value-based pharmacy management  Lower cost shares within chronic condition drug class (e.g., statins) for all individuals utilizing medication  Antihypertensives, cholesterol lowering, depression, diabetes mellitus (DM), respiratory agents (asthma/COPD)  Hypothesis: there is under-utilization of medication and all individuals utilizing medications will have a positive cost-effectiveness ratio  Goal: reduce cost as a barrier to medication initiation and persistence ●Increases in Brand ($$$$$) utilization with decrease in generic ($) utilization ●Cost-effectiveness ratio in wrong direction for non-high risk brand utilizers

8 © PRIME THERAPEUTICS LLC 8 Value-Based Insurance Outcomes ●Pitney Bowes co-insurance 10% generics, 30% formulary brand, 50% non-formulary brand  Value-based benefit design reduced Advair from 50% co-insurance (~$60/mo) to 10% (~$12/mo)  Diabetes mellitus (DM) drugs all moved to 10% co-insurance resulted in increased pharmacy expenditures, increased adherence and decreases in medical expenditures compared to benchmark data 1 ●Chernew and colleagues compared medication adherence for a large employer who implemented a $0/$12.50/$22.50 copay value benefit reduced from $5/$25/$45 compared to another large employer who did not change benefit design. 2  Adherence increased an absolute 3% antihypertensives, 4% DM, 3% statins, and 2% inhaled steroids 1) Mahoney JJ. Reducing Patient Drug Acquisition Costs Can Lower Diabetes Health Claims. Am J Manag Care 2005;11:S170-S176 2) Chernew ME, et al. Impact of Decreasing Copayments on Medication Adherence within a Disease Management Environment. Health Affairs 2008;27:103-112.

9 © PRIME THERAPEUTICS LLC 9 Value-Based Insurance Costs ●Estimated cost of moving from a $5/$25/$45 to a $0/$12.50/$22.50 copay for statins, ACE inhibitors/ARBs (antihypertensives), and asthma medications for 100,000 lives 1 :  $2.6 million to $3.9 million cost to employer  For statins, the increase in annual cost per patient would be $114 to $168 ($1.26 million to $1.86 million for 11,100 patients) to add 12.4 days of statin therapy per year 2 (absolute increase in adherence by medication possession ratio calculation of 3.39 x 365 days) 1)Fairman KA, Curtiss FR. Making the World Safe for Evidence-Based Policy: Let’s Slay the Biases in Research on Value-Based Insurance Design. J Manag Care Pharm 2008;14:198-204. 2) Chernew ME, et al. Impact of Decreasing Copayments on Medication Adherence within a Disease Management Environment. Health Affairs 2008;27:103-112.

10 © PRIME THERAPEUTICS LLC 10 Prime’s Pharmacy Benefit Trend For a copy of Prime’s Drug Trend Report, please visit www.primetherapeutics.com under “Recent News” Header click on Drug Trend Insightswww.primetherapeutics.com

11 © PRIME THERAPEUTICS LLC 11 Prime’s Core PBM Metric Trends ParameterMedianMean Total Cost PMPM$56.83$60.26 Trend6.1%2.9% Plan Cost PMPM$41.99$44.11 Trend7.1%2.6% Member Cost PMPM$15.53$16.15 Trend3.4%3.8% Utilization Change4.3%2.0% Inflation2.7%3.4% Drug Mix-0.6%-2.2%

12 © PRIME THERAPEUTICS LLC 12 Prime’s 2007 Utilization & Ingredient Cost Trend Utilization increased 2.0% Overall inflation 3.4% Note: 10.4% Brand inflation

13 © PRIME THERAPEUTICS LLC 13 2007 Drug Spend by Core Therapeutic Category Cholesterol Hypertension Depression Gastrointestinal Disorders Diabetes Pain Respiratory Women’s’ Health Seizure Disorders ADHD Allergies Psychosis Insomnia Lifestyle Heart & Vascular Percent of Prime’s clients’ drug spend Efficiency therapeutic category Non-efficiency therapeutic category

14 © PRIME THERAPEUTICS LLC 14 Clinical programs Benefit design Core Therapeutic Category Efficiency Report Value-Based Pharmacy Management Predictive Modeling: Value-Based Pharmacy Management

15 © PRIME THERAPEUTICS LLC 15 Value-Based Pharmacy Management: Predictive Modeling ●Pharmacy and medical data are integrated and predictive modeling logic is applied to identify high-risk members ●High-risk members are at serious medical event risk based on published medical literature

16 © PRIME THERAPEUTICS LLC 16 What is Done with the Predictive Modeling Information? ●High risk members pharmacy spend used to calculate efficiency metric (reporting tool) ●Identified high risk members who are not receiving optimal pharmacotherapy ●High risk members loaded into pharmacy benefit:  Lowering of member cost-share within specific drug therapeutic classes  By pass utilization management programs (e.g., preload a prior authorization for branded statins)  Adherence reporting

17 © PRIME THERAPEUTICS LLC 17 Predictive Modeling Drives Efficiency Goal The Efficiency Ratio displays drug spend for high-risk members in comparison to the spend for all members within the particular drug class. This metric is used as an indicator of how well your pharmacy dollars are being spent within each drug class. Within Pharmacy Drug Class

18 © PRIME THERAPEUTICS LLC 18 Predictive Modeling Rules ●Meta-analyses / RCTs demonstrating drug class effectiveness operationalized and guidelines used to define ICD-9 codes identifying high risk  Absolute difference in events (NNT) from publications used to estimate event reduction ●Medical claim primary field diagnosis in past two years ●Medication use defined as claim within drug class during past quarter ●Projected savings using DRGs / disability ●Risk status is permanent (not varying)

19 © PRIME THERAPEUTICS LLC 19 Predictive Modeling Efficiency Reports ●Core Areas  High cholesterol  Depression  Hypertension  Respiratory disorders  Gastrointestinal disorders  Diabetes ●Quarterly Efficiency Reports ●Adherence Reporting

20 © PRIME THERAPEUTICS LLC 20 Cholesterol Lowering Efficiency Report Employer

21 © PRIME THERAPEUTICS LLC 21 Respiratory Efficiency Report Employer

22 © PRIME THERAPEUTICS LLC 22 Gastrointestinal Efficiency Report Employer

23 © PRIME THERAPEUTICS LLC 23 How are the Efficiency Goals Met? All others

24 © PRIME THERAPEUTICS LLC 24 Aggregating All 4Q07 Reports Low (10% High Risk Increase) Medium (30% High Risk Increase) High (60% High Risk Increase) Numerator$778,907$825,432$918,483$1,058,059 Denominator$2,561,677$2,608,202$2,701,253$2,840,829 Efficiency Rating30%32%34%37% Medical SavingsN/A$203,000$789,600$1,473,000 Pharmacy SavingsN/A-$46,525-$139,576-$279,152 Efficiency Value Modeling No Change in Generic Utilization Rate (no pharmacy benefit generic utilization rate optimization) *Medical Savings are from increasing utilization among high risk members by 10% (low), 30% (medium) and 60% (high). The increased utilization among high risk projected to reduce medical events. The medical event costs estimated using CMS DRG costs. Projected annual medical savings is based upon anticipated value-based program pharmacy impact among high risk members.

25 © PRIME THERAPEUTICS LLC 25 Efficiency Value Modeling *Medical Savings are from increasing utilization among high risk members by 10% (low), 30% (medium) and 60% (high). The increased utilization among high risk projected to reduce medical events. The medical event costs estimated using CMS DRG costs. Projected annual medical savings is based upon anticipated value-based program pharmacy impact among high risk members. † Savings due to GUR rate increases at low, medium, or high rates. High GUR rates are achievable with UM and low cost formulary. Aggregating All 4Q07 Reports Low (10% HR & GUR Increase) Medium (30% HR & GUR Increase) High (60% HR & GUR Increase) Numerator$778,907$825,432$918,483$1,058,059 Denominator$2,561,677$2,400,497$2,335,251$2,260,519 Efficiency Rating30%34%39%47% Medical Savings*N/A$203,000$789,600$1,473,000 Pharmacy Savings†N/A$161,180$226,426$301,158 Implementation of pharmacy benefit optimizing generic utilization rate (GUR)

26 © PRIME THERAPEUTICS LLC 26 Standard Adherence Approach ●Standard approach looks at all members, regardless of risk ●The most common adherence measure is MPR (medication possession ratio); MPR of less than 80% is considered non- adherent Standard approach requires members to have at least one pharmacy claim to calculate adherence.

27 © PRIME THERAPEUTICS LLC 27 Efficiency Program Adherence Approach ●Efficiency Program adherence report:  Focuses on high-risk members  Targets non-adherent high-risk members  Also targets untreated high-risk members This group is ignored by the standard approach.

28 © PRIME THERAPEUTICS LLC 28 ●Standard adherence reporting looks at all members, regardless of risk, and targets non-adherent members ●Efficiency Program adherence report:  Focuses on high-risk members  Targets non-adherent high- risk members  Also targets untreated high- risk members Adherence Reporting with Efficiency Program Number of Members MPR Efficiency Program Adherence Report (high-risk members) The Efficiency Program adherence report identifies high-risk members that are untreated, or 0% adherent. These members would have been ignored by standard adherence reporting. Standard Adherence Report (all utilizers ) High Risk Untreated *Reports based on data from a 12,000 member employer group Adherent

29 © PRIME THERAPEUTICS LLC 29 Value-Based Pharmacy Management ●Targeted member interventions  High-risk members are coached via letters and Web presentations to encourage them to continue taking their medication or initiate therapy  Data on high-risk members sent to plan’s disease state management program (if applicable) for further intervention  All other members are sent letters that encourage generic utilization to lower costs ●Physician outreach to encourage treatment of untreated high-risk ●Utilization management clinical programs aimed at increasing generic utilization to manage the denominator  Step therapy  Prior authorization  Quantity limits

30 © PRIME THERAPEUTICS LLC 30 Case Study: Value-Based Benefit Design ●Set cost-sharing based on value and risk status ●Provide low cost drug therapy to members in select core therapeutic classes ●Prime recommends benefit designs that use lower out-of-pocket costs for all generic drugs and higher out-of-pocket costs for all non-formulary brand-name drugs Copay per 30-day supply Current PPO BenefitValue-based Benefit Generic25% coinsurance ($10 min & $25 max) $4 Formulary brand25% coinsurance ($20 min & $60 max) $20 Non-formulary brand25% coinsurance ($50 min & $100 max) $60 or 50% coinsurance

31 © PRIME THERAPEUTICS LLC 31 Efficiency Cholesterol Lowering Pilot Method to manage the numerator: Of 48,000 lives, 2207 lives (4.8%) were high risk Of 2207 high risk lives, 1491 (67.6%) did not have cholesterol lowering therapy High risk members not on therapy were sent educational information regarding the benefits of cholesterol lowering medications Providers of high risk members not on therapy were sent educational information identifying patients not on therapy All non-adherent members and providers were sent educational information regarding the value of medication compliance Method to manage the denominator: Step therapy was implemented to encourage the use of a generic cholesterol lowering agent before a brand

32 © PRIME THERAPEUTICS LLC 32 Efficiency Program Pilot Results Cholesterol Category Summary 2007 vs 2006 MetricRange BenchComments 25th Percentile75th Percentile % Ingredient Spend 9.0% Total Spend in the cholesterol category is much lower than Prime’s Book of Business Spend is low secondary to a younger population combined with increased use of generics PMPM Trend -7.7% Very low trend in category secondary to results from formulary changes and the implementation of cholesterol step therapy GUR 27.3% High generic utilization secondary to programs to drive generic utilization in his category Generic rate is much higher than Prime Book of business Inflation -6.2% Large decrease in inflation secondary to shift to generic utlization and resulting price drops of simvastatin primarily Utilization Change 10.1% Large increase in utilization vs 2006 possibly do to EfficiencyProgram as well as the increase in GUR resulting in lower member out of pocket costs per script improving overall compliance Mix Effect -6.5% Very low mix effect secondary to shift to use of generic products resulting from formulary change and step therapy implementation 6.1 7.710.2 38.9 24.332.5 12.5 5.114.5 -13.6 -8.2-4.8 -14.2 -16.2 -10.5-0.5 -10.8 Current data period: 01/01/2007-12/31/2007Previous data period 01/01/2006-12/31/2006Run date: 02/28/2008

33 © PRIME THERAPEUTICS LLC 33 Efficiency Program Results ●The member and provider interventions around gaps in care and adherence produced a 13% increase in utilization (1Q07 vs. 1Q08) ●Generic utilization for targeted drug category increased 15 percentage points (1Q07 vs. 1Q08) which is 12 percentage points better than Prime’s BOB benchmark for employer groups ●For every 17 high risk members intervened upon an additional 1 initiated therapy for targeted drugs ●The efficiency program goals were met:  Increased utilization  Decreased costs ●Efficiency ratio improved from 17.0% to 18.5% Next Step is to implement high risk member value-based benefit and preload statin step-therapy prior authorizations

34 © PRIME THERAPEUTICS LLC 34 Predictive Modeling Enhances Value-Based Pharmacy Management ●Understanding of how efficiently your money is spent within key drug classes ●Increase quality of care  Identify untreated high-risk members and start them on appropriate therapy for condition  Maintain adherence among high-risk users  Addressing known safety concerns by identifying any contraindications to therapy ●Maintain affordability of pharmacy benefit  Lower costs for high risk members, not all members  Apply utilization management programs to optimize generics allowing high risk members to pass through

35 © PRIME THERAPEUTICS LLC 35 Summary ●Predictive modeling helps you achieve your pharmacy management goals by:  Looking at plan pharmacy spending through the lens of medical data, stratifying members by risk  Providing a metric for trend analyses and benchmarks  Providing actionable information to make better pharmacy benefit and clinical program decisions  Giving you the information to improve the efficiency of your pharmacy dollar

36 © PRIME THERAPEUTICS LLC CONFIDENTIAL MATERIAL Pat Gleason, PharmD, BCPS, FCCP Director of Outcomes Assessment 651.414.4190 pgleason@primetherapeutics.com


Download ppt "Using Predictive Modeling to Target Value-Based Prescription Management and Pharmacy Benefit Efficiency Pat Gleason, PharmD, BCPS, FCCP Director of Outcomes."

Similar presentations


Ads by Google